Data redundancy in a hot pluggable, large symmetric multi-processor system
View Patent ↗A computer system includes a plurality of field replaceable units, each having volatile memory and at least one CPU. The FRUs communicate with each other via centralized logic. A RAID data fault tolerance technique is applied to the system so that an FRU can be lost or removed without loss of its data. An exclusive OR engine is included in the centralized logic or distributed among the FRUs. The RAID logic can restripe itself upon removal or addition of a FRU.
1. A system, comprising:
three or more volatile memory devices;
wherein data is striped across said three or more volatile memory devices to provide data redundancy; and
wherein, upon cessation of operation of a volatile memory device, said data is re-striped across the remaining volatile memory devices to continue to provide data redundancy.
2. The system of claim 1 wherein said data is striped across said three or more volatile memory devices in a first redundant array of independent disk (RAID) configuration and wherein, upon cessation of operation of the volatile memory device, said data is re-striped across the remaining volatile memory devices in a second RAID configuration.
3. The system of claim 1 wherein, upon installation of an additional volatile memory device in said system, said data is again re-striped across said volatile memory devices including said additional volatile memory device.
4. The system of claim 1 comprising a plurality of field replaceable units (FRUs), each FRU comprises at least one of said three or more volatile memory devices and an exclusive-OR engine, and each FRU is removable from said system independent from another FRU.
5. The system of claim 1 further comprising logic coupled to said three or more memory devices, and while said data is re-striped across the remaining volatile memory devices, said logic responds to a write request containing write data by computing an exclusive-OR operation on said write data and other data in said remaining volatile memory devices.
6. The system of claim 1 further comprising logic coupled to said three or more memory devices, and while said data is re-striped across the remaining volatile memory devices, said logic responds to a write request by computing an exclusive-OR operation on data associated with the requested read.
7. A system, comprising:
a plurality of volatile memory devices;
wherein data is striped across said plurality of volatile memory devices to provided data redundancy; and
wherein, upon an addition of a volatile memory device to said system, said data is re-striped across the volatile memory devices, including the newly added volatile memory device, to continue to provide data redundancy.
8. The system of claim 7 wherein said data is striped across said plurality of volatile memory devices in a first redundant array of independent disk (RAID) configuration and wherein, upon addition of the volatile memory device, said data is re-striped across the volatile memory devices in a second RAID configuration.
9. The system of claim 7 comprising a plurality of field replaceable units (FRUs), each FRU comprises at least one of said volatile memory devices and an exclusive-OR engine, and each FRU is removable from said system independent from another FRU.
10. A system, comprising:
means for configuring three or more volatile memory devices according to a first redundant array of independent disk (RAID) configuration; and
means for re-configuring the remaining volatile memory devices according to a second RAID configuration upon cessation of correct operation at least one of said three or more volatile memory devices.
11. The system of claim 10 further comprising means for re-configuring said three or more volatile memory devices, including said newly added volatile memory device, according to the first RAID configuration upon adding a volatile memory device.
12. A method, comprising:
configuring a three or more volatile memory devices according to a first redundant array of independent disk (RAID) configuration; and
upon cessation of correct operation at least one of said three or more volatile memory devices, re-configuring the remaining volatile memory devices according to a second RAID configuration.
13. The method of claim 12 further comprising adding a volatile memory device and again re-configuring said three or more volatile memory devices, including said newly added volatile memory device according to the first RAID configuration.
14. The method of claim 12 wherein configuring said three or more volatile memory devices comprises striping data across said plurality of volatile memory devices.
15. The method of claim 12 wherein re-configuring the remaining volatile memory devices comprises re-striping data across said remaining volatile memory devices.
16. The method of claim 12 further comprising performing a read request while re-configuring the remaining volatile memory devices by computing an exclusive-OR operation on data associated with the requested read.
17. The method of claim 12 further comprising, while re-configuring the remaining volatile memory devices, performing a write request, containing write data, by computing an exclusive-OR operation with said write data and other data in said remaining volatile memory devices.
18. A method, comprising:
configuring a plurality of volatile memory devices according to a first redundant array of independent disk (RAID) configuration; and
upon addition of a volatile memory device, re-configuring the volatile memory devices, including the added volatile memory device according to a second RAID configuration.
19. The method of claim 18 wherein configuring the volatile memory devices comprises striping data across said volatile memory devices.
20. The method of claim 18 wherein re-configuring the volatile memory devices comprises re-striping data across the volatile memory devices, including the added volatile memory device.